Latest

Samsung and SK Hynix just refused to prepay $18.7B of AI power bills

The Arena

Samsung and SK Hynix just refused to prepay $18.7B of AI power bills

The pacing debate has been argued in Washington and Tokyo for three days. On Monday the most useful verdict came from neither: it came from two chipmakers looking at a five-year electricity bill and saying no. Samsung Electronics and SK Hynix both rejected Korea Electric Power Corp.'s request to prepay roughly 25 trillion won (about $18.7 billion) of electricity, citing doubt over whether the AI chip boom holds. Industry sources told Reuters that KEPCO had asked Samsung for 20 trillion won and

A Guangzhou bank just lent against a compute bill, not a building

The Arena

A Guangzhou bank just lent against a compute bill, not a building

China spent two years building a token economy at industrial scale. This week it started building token-backed credit on top of it — while Washington is still arguing about who's allowed to referee the labs. China's first "token loan" landed in Guangzhou, where an API call log now does the job of a property deed. Bank of China's Haizhu branch approved a three-year, 3 million yuan unsecured loan for Tengyuan Digital, a digital-marketing firm whose founder Luo Ruichang said banks previously deman

Zhipu spoils GLM-6.0: a fully self-training model, leaked by a filing

The Frontier

Zhipu spoils GLM-6.0: a fully self-training model, leaked by a filing

The model isn't out and there's no paper yet — but Zhipu's latest funding announcement just put the company's next technical bet in writing, and it's the most explicit commitment to recursive self-improvement any frontier lab has made in a financial disclosure. Zhipu says roughly 60% of its new HK$3.93 billion raise — about HK$23.5 billion — will fund GLM-6.0 and a "fully self training" system, a term the company defines in the announcement as a closed loop where the model generates its own dat

NSA splits itself into five mission centers — one just for AI

The Guardrails

NSA splits itself into five mission centers — one just for AI

The AI pacing fight dominated Sunday's headlines. In the same news cycle, the US government quietly reorganized its own. The National Security Agency is tearing down its directorate structure and rebuilding around five "mission centers" — China, cybersecurity, artificial intelligence, combat support and global intelligence — with a 30-day clock on implementation. Washington Post reporters Zach Dorfman, Warren P. Strobel and Noah Robertson first reported the plan, sourced to current and former o

EXAONE Omni Inspect checks factory quality without retraining

The Arena

EXAONE Omni Inspect checks factory quality without retraining

LG put its industrial AI stack on stage in Seoul this morning, and the interesting part isn't the model names — it's that the inspection model no longer needs to be retrained when the product changes. Meanwhile Chinese banks started underwriting AI companies on the strength of how many tokens they burn. LG AI Research unveiled "expert AI" for manufacturing, science and finance at its AI Talk Concert 2026 today at LG Science Park in Seoul. The headline release is EXAONE Omni Inspect, an autonomo

UBTECH switches on a factory that builds a humanoid every 10 minutes

The Arena

UBTECH switches on a factory that builds a humanoid every 10 minutes

Physical AI's bottleneck just moved: China can now print humanoids at ten-thousand-unit scale, and the honest question is who's buying them. UBTECH Robotics has commissioned the world's first humanoid production line designed for ten-thousand-unit annual capacity, in Liuzhou, Guangxi. The "super smart factory" covers 14,000 square meters and is engineered to finish one industrial humanoid every 10 minutes, building the Walker S and Cruzr lines. The unusual partner is Siemens: the two co-built t

Nikkei breaks 63,000 as the AI slowdown debate hits the market

The Guardrails

Nikkei breaks 63,000 as the AI slowdown debate hits the market

The pacing debate spent the weekend arguing with Washington. On Monday morning it argued with Tokyo, and Tokyo blinked first. Tokyo delivered the first market verdict on the AI slowdown debate, and it was negative. The Nikkei 225 fell below 63,000 during Monday's session — the first time since August 4 — with AI and semiconductor names leading the selling, and SoftBank Group, Kioxia, Fujikura and Furukawa Electric among the heaviest drags. The trigger is mixed: a Saudi pipeline attack that halt

China just told AI customer service it cannot promise you a refund

The Guardrails

China just told AI customer service it cannot promise you a refund

China spent the week writing rules for the AI it already runs — who has to answer you, what the state gets to see, and where the power comes from. All three landed within days of each other, and all three are the kind of plumbing decisions that outlast model releases. China's first national standard for AI customer service, GB/T 47746—2026, took effect on September 1, and it targets the two tricks consumers hate most. The "transfer to a human" button must be easy to find, handoffs must carry th

Nadella backs AI pacing, and picks a fight over who referees

The Guardrails

Nadella backs AI pacing, and picks a fight over who referees

The pacing camp got its fourth member tonight, and the newest one arrived with an objection attached: Microsoft's CEO agrees the frontier should slow down, and does not agree that Anthropic, OpenAI and Google should be the ones deciding what that means. Satya Nadella wrote on X shortly after 7:30 PM ET that Microsoft "welcome[s] the research, focus, and deliberate pacing needed to get alignment right as the design goal," and announced a "Code of Conduct" governing Microsoft's own first-party MA

More molecules, same failures: the crack in AI drug design

The Arena

More molecules, same failures: the crack in AI drug design

China's AI-for-drug-discovery crowd spent Sunday at the Pujiang Innovation Forum arguing about the thing benchmarks don't measure — what happens after the model proposes. And the country's biggest banks are quietly building a lending product around the same question: is AI usage real enough to lend against? "The more you generate, the more you fail." That was Ding Sheng, founding dean of Tsinghua's school of pharmaceutical sciences, at the forum's Future Medicine track on September 13, and the

Claude Fable cracked a 1653 cipher. The math says it didn't

The Guardrails

Claude Fable cracked a 1653 cipher. The math says it didn't

A frontier model was handed a 370-year-old unsolved cipher and came back with a plaintext. The internet checked the work, and it doesn't hold — which makes the retraction more interesting than the claim. Vals AI announced that Claude Fable 5.1 solved the Cyphral Distich, a cryptogram of two lines of 32 numbers printed at the end of Thomas Urquhart's 1653 treatise Logopandecteision, and ranked 28th on historian Klaus Schmeh's list of the fifty most famous unsolved messages. The company's account

Today in AI — September 13, 2026

The Guardrails

Today in AI — September 13, 2026

The pacing debate ended the day where it started the week — with the labs behaving like a government. Anthropic, OpenAI and Google have been meeting since July on their own standards body, Washington's answer from the Sunday shows was that Congress would rather not, and Beijing spent the day writing doctrine around two American model names. In between: a $5 billion Chinese raise, a Nasdaq date, a honeypot that caught the "most aligned" model cheating ten times out of ten, and an investor with a

See all